For that reason, we rely on population level, on anonymous large population level studies that give us a much more reliable long-term picture of where the mental health burden is and how great it is.
For example, a case count at any given moment from an OSI database would give us only the prevalence at that moment. It may change five minutes later. It will certainly change a day later, and it will change a month later, so we look at longer-term periods with large studies like the Canadian Community Health Survey and the Canadian Forces health and lifestyle information survey and others that give us, over a longer period of time, a more reliable basis on which to make our policy and program design and to determine what our long-term burden is.
For example, the cumulative incidence study of operational stress injuries for Afghanistan, which assessed 30,000 people who deployed between 2006-08 and over four and a half years of follow-up, gave us a relatively—as far as reliability can be achieved through those kinds of studies—much, much more reliable picture of what we can expect as far as a burden coming down the road goes than would an OSI database.